A heterogeneous resource evaluation method and system based on dynamic behavior feature clustering

Abstract

The application belongs to the technical field of power system operation and control, and provides a heterogeneous resource evaluation method and system based on dynamic behavior feature clustering, obtains resource grid-connected point electrical measurement data and regional meteorological environment data, constructs a multi-dimensional time series data set containing electrical quantity and environmental quantity; according to the multi-dimensional time series data set, multi-dimensional electrical behavior features of each resource are extracted; according to the multi-dimensional electrical behavior features, the coupling relationship of active power and reactive power under different environmental time sequence scenes is analyzed, a clustering algorithm is used to classify and divide the resources; a multi-dimensional evaluation model is established according to the classified resources, the multi-dimensional regulation capability of different types of resources in the electrical aspect is comprehensively evaluated, a radar chart is generated, and according to the shape of the radar chart, the optimal resource combination is selected in combination with the state of the power grid. The application can realize the quantitative resource comprehensive capability in the dimensions of active power balance, voltage support, frequency response, etc., so as to better accurately schedule the resources according to the state of the power grid.

Description

Technical Field

[0001] This invention belongs to the field of power system operation and control technology, specifically relating to a heterogeneous resource assessment method and system based on dynamic behavioral feature clustering. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] With the development of new power systems, users are no longer simply loads, but rather producers and consumers integrating various devices such as distributed photovoltaics, energy storage, and electric vehicles. Simultaneously, the interaction between the random volatility of new energy sources and the uncertainty of electricity user behavior leads to increasingly complex distribution network operation, severely impacting the accuracy of traditional load assessment and adjustable resource assessment. In new power systems, the source-load boundary is blurred, but the physical foundation for maintaining system stability remains unchanged: the constant maintenance of active power-frequency balance and reactive power-voltage balance. Active power and frequency are the benchmarks for measuring the system's dynamic balancing capability, while reactive power and voltage are the cornerstones for measuring the system's static / transient support capability. Quantitative assessment of these four dimensions can effectively capture resource adjustment potential and response capabilities from high-frequency measurement data, avoiding interference from secondary factors in capability assessment.

[0004] However, existing technologies mostly employ resource classification methods based on single factors such as time or temperature sensitivity, making it difficult to accurately capture and quantify user behavior differences and adjustability potential during special periods. For example, air conditioning load and energy storage are similar in meteorological sensitivity, but their value in grid ancillary services is completely different. At the same time, they only focus on the prediction error of active power and lack a comprehensive assessment of reactive power, frequency and voltage support capabilities, which cannot support multiple types of transaction decisions such as voltage and frequency regulation. Traditional static label classification also ignores the dynamic adjustment potential of resources under different operating conditions. Summary of the Invention

[0005] To address the aforementioned issues, this invention proposes a heterogeneous resource assessment method and system based on dynamic behavioral feature clustering. This invention addresses the differences in electrical response characteristics and environmental impact among various resource types in measurement data. Based on features such as ramp rate, PQ coupling, response dead zone, and environmental sensitivity, it employs unsupervised clustering technology to identify resource types and quantify their comprehensive capabilities in dimensions such as active power balance, voltage support, and frequency response, thereby enabling more accurate resource scheduling based on grid conditions.

[0006] According to some embodiments, the present invention adopts the following technical solution: A heterogeneous resource assessment method based on dynamic behavioral feature clustering includes: Acquire electrical measurement data from resource grid connection points and regional meteorological and environmental data, and construct a multi-dimensional time-series dataset containing electrical and environmental quantities; Based on the multidimensional time-series dataset, extract multidimensional electrical behavior features of each resource; Based on the multidimensional electrical behavior characteristics, the coupling relationship between active and reactive power under different environmental time-series scenarios is analyzed. Clustering algorithms are used to classify resources into flexible resources, capacity resources and rigid resources. A multidimensional evaluation model is established based on the classified resources to comprehensively evaluate the multidimensional adjustment capabilities of different types of resources in the electrical aspect, and a resource capability radar chart is generated to quantify the contribution of resources in terms of active power, voltage, frequency and reliability. Based on the shape of the radar chart and the state of the power grid, select the optimal resource combination.

[0007] As an optional implementation method, the process of acquiring electrical measurement data of the resource grid connection point and regional meteorological environment data includes: acquiring active power, reactive power, voltage, and frequency data of the resource grid connection point, and acquiring temperature and light intensity data of the resource grid connection point area.

[0008] As an alternative implementation method, the process of constructing a multidimensional time-series dataset containing electrical quantities and environmental quantities includes: preprocessing the acquired electrical measurement data and regional meteorological and environmental data, using the regional meteorological and environmental data as environmental quantities and as independent variables, and using the electrical measurement data as electrical quantities and as dependent variables, and constructing the dataset through time-series relationships.

[0009] As an alternative implementation, the multidimensional electrical behavior characteristics include dynamic ramp rate, active-reactive coupling degree, peak-valley response contribution, fluctuation uncertainty, and temperature-active time-series lag correlation.

[0010] As a further defined implementation, the dynamic ramp rate is obtained by calculating the absolute average of power changes, and is used to characterize the potential of resource tracking automatic generation control commands or primary frequency regulation. The active-reactive coupling degree is obtained by calculating the correlation coefficient between active power and reactive power. If the correlation coefficient is less than a preset threshold, it is used to characterize that the resource has the ability to decouple active and reactive power. The peak-valley response contribution is used to characterize the willingness of resources to shift or reduce load during peak periods of the power grid by calculating power deviation. The fluctuation uncertainty is used to measure the randomness of resource output, serving as a negative indicator for reliability assessment; The temperature-active power time-series lag correlation takes into account the time lag. By calculating the cross-correlation function between active power and temperature, it is used to characterize the resource active power regulation capability considering the time lag.

[0011] As an alternative implementation method, the process of classifying resources into flexible resources, capacity resources, and rigid resources using a clustering algorithm includes: normalizing multidimensional electrical behavior characteristics and using a clustering algorithm to classify them into flexible resources (dynamic ramp rate higher than a set value, active and reactive power coupling degree lower than a set threshold, suitable for frequency regulation and voltage support); capacity resources (dynamic ramp rate lower than a set value, peak-valley response contribution higher than a set threshold, suitable for peak shaving and valley filling); and rigid resources (fluctuation uncertainty less than a set value, dynamic ramp rate less than a set value, peak-valley response contribution less than a set threshold, serving as baseline load).

[0012] As an alternative implementation method, the process of establishing a multi-dimensional evaluation model based on the classified resources and comprehensively evaluating the multi-dimensional adjustment capabilities of different types of resources in terms of electricity includes: calculating the active power regulation capability index based on peak-valley response contribution and maximum adjustable capacity. The voltage support capability index is calculated based on the active and reactive power decoupling characteristics and the apparent power remaining capacity of the inverter. The frequency response capability index is proportional to the dynamic ramp rate; The reliability index is inversely proportional to the volatility and uncertainty characteristics; Using the four indices above as evaluation indicators, assess the value of each index for each resource, and draw a radar chart of the overall score of the resource based on the values.

[0013] As an alternative implementation method, the process of selecting the optimal resource combination based on the shape of the radar chart and the power grid status includes: when the voltage exceeds the limit, resources with a voltage support capability index greater than the set value are prioritized; when the frequency drops, resources with a frequency response capability index greater than the set value are prioritized.

[0014] A heterogeneous resource assessment system based on dynamic behavioral feature clustering includes: The information acquisition module is configured to acquire electrical measurement data of resource grid connection points and regional meteorological and environmental data, and construct a multi-dimensional time-series dataset containing electrical quantities and environmental quantities. The multidimensional electrical behavior feature extraction module is configured to extract multidimensional electrical behavior features of each resource based on a multidimensional time-series dataset. The heterogeneous resource classification module is configured to analyze the coupling relationship between active and reactive power in different environmental time-series scenarios based on multidimensional electrical behavior characteristics, and classify resources into flexible resources, capacity resources and rigid resources using clustering algorithms. The multidimensional regulation capability comprehensive evaluation module is configured to establish a multidimensional evaluation model based on the classified resources, comprehensively evaluate the multidimensional regulation capability of different types of resources in the electrical aspect, and generate a resource capability radar chart to quantify the contribution of resources in terms of active power, voltage, frequency and reliability. The resource scheduling module is configured to select the optimal resource combination based on the shape of the radar chart and the power grid status.

[0015] An electronic device includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, wherein the computer instructions, when executed by the processor, perform the steps in the method described above.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: The multi-dimensional regulation capability assessment system for heterogeneous resources based on dynamic behavior clustering provided by this invention tightly integrates active and reactive power coupling with climate time-series characteristics. It no longer relies on unreliable static labels, but instead extracts features directly from electrical behavior through a data-driven approach, achieving a precise profile of massive, dispersed resources. Through active and reactive power coupling analysis, it effectively identifies power electronic interface devices with reactive power support capabilities, such as photovoltaic inverters and energy storage. Through multi-dimensional radar chart scoring, it achieves precise matching of resources with the needs of peak-shaving, voltage-regulating, and frequency-regulating power grids, improving the utilization efficiency and operational reliability of the distribution network for heterogeneous resources.

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0018] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0019] Figure 1 This is a flowchart of a heterogeneous resource evaluation method based on dynamic behavioral feature clustering in Embodiment 1 of the present invention; Figure 2 This is a flowchart of heterogeneous resource clustering based on behavioral dynamic features in Embodiment 1 of the present invention; Figure 3 This is the result of multi-type resource aggregation in Embodiment 1 of the present invention, which comprehensively considers the active and reactive power coupling degree and the ramp rate. Figure 4 The results are the assessment results of the adjustment capabilities of different types of resources in Embodiment 1 of the present invention. Detailed Implementation

[0020] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0021] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0022] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0023] Where there is no conflict, the embodiments and features described in this application may be combined with each other.

[0024] Example 1 Embodiment 1 of this invention introduces a method for evaluating the multidimensional adjustment capability of heterogeneous resources based on dynamic behavior clustering, such as... Figure 1 As shown, it includes: Acquire measurement data such as active power, reactive power, voltage, and frequency at the resource grid connection point, as well as regional meteorological and environmental data, and construct a multi-dimensional time-series dataset containing electrical and environmental quantities; Based on the multidimensional time-series dataset, we construct multidimensional indicators to reflect the physical regulation capability of resources and extract multidimensional electrical behavior features of different types of resources. Based on resource characteristics, the coupling relationship between active and reactive power under different environmental time-series scenarios is analyzed, and a clustering algorithm is used to process the data, classifying resources into flexible resources, capacity resources, and rigid resources. A multi-dimensional capability assessment model is established based on the classified resources to comprehensively evaluate the multi-dimensional regulation capabilities of different types of resources in terms of active power, reactive power, frequency, and voltage, and to generate a resource capability radar chart to quantify their contribution to active power, voltage, frequency, and reliability. Based on the shape of the radar chart and the state of the power grid, select the optimal resource combination.

[0025] In this embodiment, the active power, reactive power, voltage, frequency, temperature, and illuminance data involve multiple time scales based on peak shaving, voltage regulation, and frequency regulation requirements. To construct a time-series dataset of these data, this embodiment synchronously collects grid connection point frequency and active power data at a frequency of no less than 1Hz to capture resource frequency regulation response characteristics. To accurately construct the environmentally constrained dynamic coupling characteristics of active and reactive power, active power, reactive power, voltage, and illuminance data are collected at a frequency of 1 second / time to capture voltage support behavior during cloud cover. Considering the thermal inertia characteristics of temperature-controlled resources, active power and temperature data are collected at a frequency of 15 minutes / time.

[0026] Based on the obtained multi-timescale active and reactive power, voltage, frequency, temperature and light intensity data, this embodiment constructs multi-dimensional indicators to reflect the physical regulation capability of resources and extracts multi-dimensional electrical behavior characteristics of different types of resources.

[0027] The dynamic ramp rate, calculated by the absolute average of power changes, is used to characterize the potential for resource tracking of automatic generation control commands or primary frequency regulation. (1) In the formula, P t for t Active power at time Δ t The sampling interval is... T This represents the total number of samples. The larger this value, the faster the resource adjustment speed.

[0028] Specifically, active and reactive power coupling degree ρ PQ The active and reactive power decoupling capability of resources is characterized by calculating the covariance of active and reactive power: (2) In the formula, cov is the covariance. σ P and σ Q These are the standard deviations of active and reactive power, respectively. If ρ PQ < ε This indicates that the resource possesses active and reactive power decoupling control capabilities and can independently provide voltage support. ε A threshold close to 0; | ρ PQ | Approaching 1 indicates a traditional inductive load or undecoupled control device.

[0029] Specifically, peak-valley response contribution C pv The ability to smooth out peaks and fill valleys is characterized by calculating power deviation: (3) In the formula, T peak A set of peak periods defined for the power grid. P base For baseline load, P actual This represents the actual load.

[0030] Specifically, volatility uncertainty σ unc Measuring the randomness of resource output: (4) In the formula, P real,t for t The measured value of the actual output of resources at any given time. P pred,t for tPredicted values ​​of resource output at any given time μ error This is the average value of the prediction error. T To evaluate the total number of data sampling points within the time window.

[0031] Specifically, the temperature-active power time-series lag correlation β T Considering the time lag, the ability to provide active power is characterized by calculating the cross-correlation function between active power and temperature: (5) In the formula, T emp,t-τ To take time delay into account τ of t Temperature at all times σ Temp denoted as the standard deviation of temperature.

[0032] like Figure 2 , Figure 3 As shown, this embodiment analyzes the active and reactive power coupling relationship under different time-series scenarios through multi-dimensional electrical behavior characteristics of resources, normalizes the feature data to eliminate the influence of dimensions, and divides users into K clusters. Based on the cluster center characteristics, these clusters are defined as follows: Flexibility Resources: Cluster centers exhibit high... R amp ,Low ρ PQ Typical examples include energy storage and V2G electric vehicles. These resources are suitable for participating in grid frequency regulation and voltage support services; capacity-type resource cluster centers exhibit low... R amp ,high C pv ,high β T Typical equipment such as electric heating and water heaters are suitable for peak shaving and valley filling as well as backup services; rigid resource cluster centers are characterized by low... σ unc ,Low C pv ,Low R amp As a baseline load, it does not have significant adjustment potential.

[0033] Based on the resource classification results, this embodiment establishes a multi-dimensional evaluation model to comprehensively evaluate the multi-dimensional regulation capabilities of different types of resources in terms of active power, reactive power, frequency, and voltage.

[0034] Specifically, the active power regulation capacity index I P Based on peak and valley response contribution C pv Calculated from the maximum adjustable capacity: (6) In the formula, P rated Rated power, w These are the weighting coefficients.

[0035] Specifically, voltage support capability index I V Based on the active and reactive power decoupling characteristics and the inverter's apparent power surplus capacity, the following decoupling resources are calculated: (7) In the formula, S inv This represents the apparent power capacity of the inverter. The higher this index, the greater the reactive power support potential.

[0036] Specifically, frequency response capability index I f Proportional to dynamic ramp rate, inversely proportional to response dead time t delay .

[0037] (8) Specifically, reliability index I R It is inversely proportional to the characteristics of volatility and uncertainty.

[0038] (9) like Figure 4 As shown, this embodiment ultimately outputs a comprehensive score radar chart for each aggregated resource, and the scheduling center can select the optimal resource combination based on the shape of the radar chart.

[0039] In this embodiment, the current operating condition is identified based on real-time power grid monitoring data, and the four-dimensional capability index of various resources (active power regulation) is matched. I P Voltage support I V Frequency response I f ,reliability I R The following hierarchical optimization scheduling strategy will be implemented: Operating Condition 1: Power grid frequency exceeds limits or a step disturbance occurs (frequency emergency support condition). Power grid status criterion: When a system frequency deviation Δ is detected... f >Δ f set or rate of change of frequency |d f / d t |> εResource selection logic: Filtering: First, lock in the set classified as "flexibility resources" and eliminate those with response dead time. t delay Resources whose response time exceeds the system's allowed response time; among the filtered resources, those are ranked according to their frequency response capability index. I f Sort by size from largest to smallest; call first. I f The resource with the highest value is used for rapid active power support; if the capacity of the preferred resource is insufficient, it is added in order of priority until the power gap required for frequency recovery is met.

[0040] Operating Condition 2: Local Voltage Exceeds Limits in Distribution Network (Voltage Correction Condition) Grid Status Criteria: When the voltage deviation at the grid connection point or critical node Δ... U >Δ U set Time. Resource selection logic: Filtering: Lock all resources with active and reactive power decoupling capabilities (i.e., ρ PQ Resources below the threshold); based on the voltage support capability index. I V Sort by size from largest to smallest; prioritize scheduling. I V High-efficiency resources output reactive power to support voltage. If active power reduction is required to coordinate with voltage regulation, the "active-reactive power replacement cost" of each resource is further calculated, and the resource combination with the lowest cost is selected.

[0041] Operating Condition 3: Peak Grid Load or Line Overload (Peak Shaving and Valley Filling Condition) Grid Status Criteria: When the line load rate... > limit Or during the preset peak electricity price period T peak The set classified as "capacity-type resources" is locked in, based on the active power regulation capacity index. I P Sort from largest to smallest, the index combines peak-valley response contribution and maximum adjustable capacity, according to... I P Generate a list of peak-shaving instructions sequentially; for I P Under the same circumstances, further compare their reliability indices. I R Prioritize calling outputs with low volatility (i.e., I R High) resources to prevent the introduction of new power fluctuation risks during heavy loads.

[0042] Operating Condition 4: Insufficient Safety Margin Due to Dual Uncertainty of Source and Load (Smoothing Fluctuation Condition) Grid Status Criteria: When the prediction error of new energy sources exceeds the threshold or the system reserve capacity is insufficient, exclude all "rigid resources" and select all remaining adjustable resources; construct a comprehensive scoring function. ,in α > β Emphasizing reliability index I R Weighting; Selecting the overall score S The highest resource clusters serve as backup units or regulators to smooth out fluctuations, leveraging their low volatility uncertainty to provide stable baseline support for the power grid.

[0043] Example 2 Embodiment 2 of this invention introduces a multi-dimensional adjustment capability assessment system for heterogeneous resources based on dynamic behavioral feature clustering, comprising: The information acquisition module is configured to acquire electrical measurement data of resource grid connection points and regional meteorological and environmental data, and construct a multi-dimensional time-series dataset containing electrical quantities and environmental quantities. The multidimensional electrical behavior feature extraction module is configured to extract multidimensional electrical behavior features of each resource based on a multidimensional time-series dataset. The heterogeneous resource classification module is configured to analyze the coupling relationship between active and reactive power in different environmental time-series scenarios based on multidimensional electrical behavior characteristics, and classify resources into flexible resources, capacity resources and rigid resources using clustering algorithms. The multidimensional regulation capability comprehensive evaluation module is configured to establish a multidimensional evaluation model based on the classified resources, comprehensively evaluate the multidimensional regulation capability of different types of resources in the electrical aspect, and generate a resource capability radar chart to quantify the contribution of resources in terms of active power, voltage, frequency and reliability. The resource scheduling module is configured to select the optimal resource combination based on the shape of the radar chart and the power grid status.

[0044] The detailed steps are the same as those of the heterogeneous resource scheduling method based on dynamic behavioral feature clustering provided in Example 1, and will not be repeated here.

[0045] Example 3 Embodiment 3 of the present invention provides an electronic device.

[0046] An electronic device includes a memory, a processor, and a program stored in the memory and running on the processor. When the processor executes the program, it implements the steps in the heterogeneous resource scheduling method based on dynamic behavioral feature clustering as described in Embodiment 1 of the present invention.

[0047] The detailed steps are the same as those provided in Embodiment 1 for the heterogeneous resource scheduling method based on dynamic behavioral feature clustering, and will not be repeated here. Those skilled in the art should understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of one or more computer-usable storage media (including but not limited to disk storage, etc.) containing computer-usable program code. CD - ROM It takes the form of a computer program product implemented on (such as optical memory, etc.).

[0048] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0049] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0050] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0051] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made by those skilled in the art without creative effort within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A heterogeneous resource assessment method based on dynamic behavioral feature clustering, characterized in that, include: Acquire electrical measurement data from resource grid connection points and regional meteorological and environmental data, and construct a multi-dimensional time-series dataset containing electrical and environmental quantities; Based on the multidimensional time-series dataset, extract multidimensional electrical behavior features of each resource; Based on the multidimensional electrical behavior characteristics, the coupling relationship between active and reactive power under different environmental time-series scenarios is analyzed. Clustering algorithms are used to classify resources into flexible resources, capacity resources and rigid resources. A multidimensional evaluation model is established based on the classified resources to comprehensively evaluate the multidimensional adjustment capabilities of different types of resources in the electrical aspect, and a resource capability radar chart is generated to quantify the contribution of resources in terms of active power, voltage, frequency and reliability. Based on the shape of the radar chart and the state of the power grid, select the optimal resource combination.

2. The heterogeneous resource assessment method based on dynamic behavioral feature clustering as described in claim 1, characterized in that, The process of acquiring electrical measurement data of the resource grid connection point and regional meteorological environment data includes: acquiring active power, reactive power, voltage, and frequency data of the resource grid connection point, and acquiring temperature and light intensity data of the resource grid connection point area.

3. The heterogeneous resource assessment method based on dynamic behavioral feature clustering as described in claim 1, characterized in that, The process of constructing a multidimensional time-series dataset containing electrical quantities and environmental quantities includes: preprocessing the acquired electrical measurement data and regional meteorological and environmental data, using the regional meteorological and environmental data as environmental quantities and as independent variables, and using the electrical measurement data as electrical quantities and as dependent variables, and constructing the dataset through time-series relationships.

4. The heterogeneous resource assessment method based on dynamic behavioral feature clustering as described in claim 1, characterized in that, The multidimensional electrical behavior characteristics include dynamic ramp rate, active-reactive coupling degree, peak-valley response contribution, fluctuation uncertainty, and temperature-active time lag correlation.

5. The heterogeneous resource assessment method based on dynamic behavioral feature clustering as described in claim 4, characterized in that, The dynamic ramp rate is obtained by calculating the absolute average of power changes and is used to characterize the potential of resource tracking automatic generation control commands or primary frequency regulation. The active-reactive coupling degree is obtained by calculating the correlation coefficient between active power and reactive power. If the correlation coefficient is less than a preset threshold, it is used to characterize that the resource has the ability to decouple active and reactive power. The peak-valley response contribution is used to characterize the willingness of resources to shift or reduce load during peak periods of the power grid by calculating power deviation. The fluctuation uncertainty is used to measure the randomness of resource output, serving as a negative indicator for reliability assessment; The temperature-active power time-series lag correlation takes into account the time lag. By calculating the cross-correlation function between active power and temperature, it is used to characterize the resource active power regulation capability considering the time lag.

6. The heterogeneous resource assessment method based on dynamic behavioral feature clustering as described in claim 1, characterized in that, The process of classifying resources into flexible resources, capacity resources, and rigid resources using clustering algorithms includes: normalizing multidimensional electrical behavior characteristics and then using clustering algorithms to classify them into flexible resources (dynamic ramp rate higher than a set value, active and reactive power coupling degree lower than a set threshold, suitable for frequency regulation and voltage support); capacity resources (dynamic ramp rate lower than a set value, peak-valley response contribution higher than a set threshold, suitable for peak shaving and valley filling); and rigid resources (fluctuation uncertainty less than a set value, dynamic ramp rate less than a set value, peak-valley response contribution less than a set threshold, serving as baseline load).

7. The heterogeneous resource assessment method based on dynamic behavioral feature clustering as described in claim 1, characterized in that, The process of establishing a multidimensional evaluation model based on the classified resources and comprehensively evaluating the multidimensional adjustment capabilities of different types of resources in terms of electricity includes: calculating the active power regulation capability index based on peak-valley response contribution and maximum adjustable capacity; The voltage support capability index is calculated based on the active and reactive power decoupling characteristics and the apparent power remaining capacity of the inverter. The frequency response capability index is proportional to the dynamic ramp rate; The reliability index is inversely proportional to the volatility and uncertainty characteristics; Using the four indices above as evaluation indicators, assess the value of each index for each resource, and draw a radar chart of the overall score of the resource based on the values.

8. The heterogeneous resource assessment method based on dynamic behavioral feature clustering as described in claim 1, characterized in that, The process of selecting the optimal resource combination based on the shape of the radar chart and the power grid status specifically includes: when the power grid is detected to be in a frequency emergency support condition, screening flexible resources and sorting them from largest to smallest according to the frequency response capability index, and prioritizing the use of resources with high dynamic ramp rate and short response dead time for power support. When the grid is detected to be in voltage correction mode, resources with active and reactive power decoupling characteristics are selected and sorted from largest to smallest according to the voltage support capacity index. Resources with large apparent power remaining capacity of the inverter are given priority for reactive power compensation. When the power grid is detected to be in peak shaving and valley filling mode, capacity-type resources are screened and sorted from largest to smallest according to the active power regulation capacity index. Resources with high peak-valley response contribution are prioritized for load transfer or reduction. When the power grid is detected to be in a state of smoothing fluctuations, the comprehensive score of the resources is calculated, with the weight of increasing the reliability index, and resources with low fluctuation uncertainty are prioritized as stabilizers.

9. A heterogeneous resource assessment system based on dynamic behavioral feature clustering, characterized in that, include: The information acquisition module is configured to acquire electrical measurement data of resource grid connection points and regional meteorological and environmental data, and construct a multi-dimensional time-series dataset containing electrical quantities and environmental quantities. The multidimensional electrical behavior feature extraction module is configured to extract multidimensional electrical behavior features of each resource based on a multidimensional time-series dataset. The heterogeneous resource classification module is configured to analyze the coupling relationship between active and reactive power in different environmental time-series scenarios based on multidimensional electrical behavior characteristics, and classify resources into flexible resources, capacity resources and rigid resources using clustering algorithms. The multidimensional regulation capability comprehensive evaluation module is configured to establish a multidimensional evaluation model based on the classified resources, comprehensively evaluate the multidimensional regulation capability of different types of resources in the electrical aspect, and generate a resource capability radar chart to quantify the contribution of resources in terms of active power, voltage, frequency and reliability. The resource scheduling module is configured to select the optimal resource combination based on the shape of the radar chart and the power grid status.

10. An electronic device, characterized in that, It includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, which, when executed by the processor, perform the steps of the method according to any one of claims 1-8.

Images